Research Methods : Statistics Flashcards
What are statistics and why are they used in psychology?
- Statistics are a branch of mathematics used to analyse data.
- They’re used by psychologists to aid the process of hypothesis testing.
- Psychologists use statistics to analyse the data they collect and determine if their hypothesis is supported.
What is the difference between correlational and experimental research?
- Presence of manipulation: experiments must have one variable being manipulated (IV) and one being measured (DV). In correlational the researcher doesn’t manipulate any variables, but instead measure all the variables.
- Purpose of investigation - The aim of an experiment is to establish a cause and effect relationship . But the aim of a correlational study is to simply establish whether two or more variables are associated in some way.
- Use of controls - random allocation or counterbalancing are used in an experiment as the IV is being manipulated. Correlational research cannot use these controls.
- Presentation of data - if research is represented in a Scattergraph than it is correlational. If the research can be represented in a bar chat then the study is an experiment.
- Descriptive statistics- these describe a pattern of data. An experiment would describe patterns in the data with measures of central tendency for the DV in both conditions which would allow the researchers to compare which condition did better. A correlational study would describe the strength of the data using the correlation coefficient.
What is the alternative and null hypothesis?
- Alternative hypotheses are clear statements about the expected relationship between the variables in a research study. (Directional and non-directional)
- A null hypothesis is used when no difference or correlation can be found.
What is Nominal level of measurement?
- DEFINITION:The simplest level of measurement/least precise. Involves assigning labels or names to identify categorise objects or subjects without any quantitative value or order.
- CHARACTERISTICS: No inherent order or ranking; categories are discrete, which means mutually exclusive (I.e., you can only be in one category).
- EXAMPLE: colours, genders.
What is Ordinal level of measurement?
- DEFINITION: n: Ordinal level involves categorizing data into discrete groups with a meaningful order or ranking among the categories. However, the intervals between
the categories are not uniform (so each category is “more” than the previous, but there isn’t a specific amount that category A is “more” than category B). The
reason for this is that there is an element of subjectivity to the ordering of the data. The ordinal level of measurement is more precise than nominal, but less precise
than interval. - CHARACTERISTICS: Categories have a meaningful order, but differences between categories are not consistent.
- EXAMPLES : education level, customer satisfaction ratings.
What is Interval level of measurement?
- DEFINITION : Interval level involves numerical scales with a consistent interval or distance between values. This means that the difference .between 1-2, 2-3,5-6, or
100-101 is precisely the same. It’s therefore possible with interval data to say that a score is exactly twice as high as another sore (e.g., 2cm is exactly twice as large as 1cm, just as 100seconds is exactly twice as long as 50 seconds). Interval data is the most precise level of measurement. - EXAMPLES: temperature, years
When do you use Mean?
Appropriate only for interval data BUT do not use if the interval data is distorted by extreme values/outliers (the median is better in this situation, as it’s less affected by outliers)
How do you carry out measuring the Mean?
Add together all scores and divide by total number of scores
How do you interpret the mean?
The mean is a measure of central
tendency (I.e., it describes the average or
“central” value for a data set). Examining
the mean for the conditions in an
experiment can enable a researcher to
make an initial judgement on which
hypothesis (null or alternative) should be
accepted or rejected
When do you use the mean?
Appropriate for ordinal dataAlso appropriate for interval data. Typically, researchers use the mean for interval data, but the median should be used in situations where the data has significant outliers, (these outliers would distort the mean, whilst the median is less
affected by outliers)
How do you carry out measuring the median?
Arrange scores from lowest to hights. The middle value is the median (if there’s an even number of scores, then the median is the mean of the two middle scores)
How do you interpret the median?
The median is a measure of central
tendency (I.e., it describes the average or “central” the average value (i.e., the
central value). Examining the median for the conditions in an experiment can
enable a researcher to make an initial
judgement on which hypothesis (null or
alternative) should be accepted or
rejected
When do you use the mode?
If the level of measurement is nominal.
How do you interpret the mode?
The mode is a measure of central
tendency (I.e., it describes the average or “central” value). Examining the mode for the conditions in an experiment can
enable a researcher to make an initial
judgement on which hypothesis (null or
alternative) should be accepted or
rejected
When do you use the range?
Use for measuring the distribution of ordinal data. Can be used with interval, but the range is so imprecise that it’s much better to use standard deviation
